Methods, apparatus, electronic devices and computer-readable storage media for assessing equivalent PM2.5 exposure concentrations for urban residents

Abstract

This invention relates to a method, apparatus, electronic device, and computer-readable storage medium for assessing equivalent PM2.5 exposure concentrations of urban residents. Belonging to the field of environmental monitoring, it aims to improve the accuracy of obtaining PM2.5 exposure concentrations for urban residents. Based on annual outdoor environmental data, it obtains annual equivalent PM2.5 exposure concentration values ​​for urban residents, including a first equivalent exposure concentration value; based on annual indoor environmental data, it obtains annual equivalent PM2.5 exposure concentration values ​​for urban residents, including a second equivalent exposure concentration value; and based on the first and second equivalent exposure concentration values, it obtains a comprehensive annual equivalent PM2.5 exposure concentration value for urban residents. This invention incorporates the interaction between air pollution risk information and protective behaviors in determining urban residents' PM2.5 exposure concentrations, resulting in PM2.5 exposure concentrations that better reflect actual conditions.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring and relates to a method and device for assessing the equivalent PM2.5 exposure concentration of urban residents. Background Technology

[0002] Air pollution is one of the most serious environmental problems faced by humankind since the industrial age, imposing a heavy burden on human health and social development. Current research primarily focuses on assessing the impact of the implementation of relevant policies and standards, such as clean fuel substitution programs, integrated clean air programs, and expected air quality standards, on the harmful effects of air pollution; and the impact of individual-level air pollution exposure levels at different times and spaces on health. The former mainly focuses on the macro-level impact of air pollution on health, often neglecting differences in air pollution exposure at different times and spaces for individual activities, or the interactions between different environmental air pollution sources. The latter focuses on micro-level individual pollution exposure. A common characteristic of these studies is that they treat humans as passive subjects, meaning that the level of air pollution experienced by individuals depends solely on the level of ambient air pollution.

[0003] However, in reality, people don't passively endure the harm caused by air pollution. According to risk perception theory, when people realize that air pollution poses a threat to their health, they will take action to mitigate or eliminate this risk. This self-protective behavior alters people's activity patterns, thereby directly or indirectly reducing their exposure to polluted air. Therefore, ignoring proactive protective measures will inevitably affect the accurate assessment of the harm caused by air pollution. The government has implemented a series of related measures to improve the situation. One important change is the revision of air quality standards, which for the first time included highly harmful pollutants such as PM2.5 in the air quality assessment system. The number of pollutants monitored has increased from three to six, and local government departments are required to publish real-time air quality information to the public through various channels. As a result, air quality monitoring and early warning information is as readily available in daily life as weather information, allowing people to decide whether to take protective measures based on the published data.

[0004] Based on the aforementioned research, any research should be grounded in more accurate and comprehensive environmental pollution data. However, the PM2.5 exposure concentration information used in the current research does not incorporate the interaction between air pollution risk information and protective behaviors. Consequently, the PM2.5 exposure concentration information used in the models of these studies does not accurately reflect actual pollution information, and the research results obtained based on this data may deviate from reality. Summary of the Invention

[0005] To address the issue of improving the accuracy of obtaining PM2.5 exposure concentration data for urban residents, in a first aspect, a method for assessing equivalent PM2.5 exposure concentration for urban residents according to some embodiments of this application includes...

[0006] Based on annual outdoor environmental data of urban residents, the annual equivalent PM2.5 exposure concentration values ​​of urban residents are obtained, including the first equivalent exposure concentration value.

[0007] Based on annual indoor environmental data of urban residents, the annual equivalent PM2.5 exposure concentration values ​​of urban residents' indoor environment are obtained, including the second equivalent exposure concentration values;

[0008] Based on the first equivalent exposure concentration value and the second equivalent exposure concentration value, the annual comprehensive PM2.5 equivalent exposure concentration value for urban residents is obtained.

[0009] According to some embodiments of the present application, the method for assessing the equivalent PM2.5 exposure concentration of urban residents, the step of obtaining the first equivalent exposure concentration value includes the following steps:

[0010] Based on the proportion of daily average outdoor time spent by residents of city i, the proportion of residents of city i who cancel their outdoor activities during periods of air pollution, and the proportion of residents of city i who pay attention to air pollution forecasts, the equivalent outdoor time ratio (ODR) for city i, considering those who pay attention to air pollution forecasts, is obtained. i It is expressed by the following formula:

[0011] ODR i =ODR i *(1-CODR i *ATTR i )

[0012] Among them: ODR i CODR represents the proportion of daily outdoor time spent by residents of city i. i ATTR indicates the percentage of residents in city i who cancel their outdoor activities during periods of air pollution. i This indicates the percentage of residents in city i who pay attention to air pollution forecast information;

[0013] Based on the proportion of residents in city i who wear masks during outdoor activities during periods of air pollution and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent outdoor mask-wearing rate MR of city i is obtained. i It is expressed by the following formula:

[0014] MR i =MR i *ATTR i

[0015] Among them: MR iThis indicates the percentage of residents in city i who wear masks during outdoor activities when air pollution is present.

[0016] Based on the average PM2.5 concentration on day t of a year in city i, the proportion of people in city i who spend the equivalent time outdoors when considering air pollution forecast information, the equivalent mask-wearing rate outdoors in city i, the PM2.5 penetration rate of masks, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. AAP，i It is expressed by the following formula:

[0017]

[0018] Among them: IEPE AAP，i This represents the annual equivalent PM2.5 exposure concentration for residents of city i in the outdoor environment. i，t ODR represents the average PM2.5 concentration on day t of city i in a year. i MR represents the equivalent outdoor time proportion of people in city i when considering those who pay attention to air pollution forecast information. i DM represents the equivalent mask-wearing rate in city i, DM represents the PM2.5 penetration rate of the mask, and AQI represents the PM2.5 penetration rate. i，t Let represent the average air quality index (AQI) value of city i on day t of a year, where T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is lightly polluted.

[0019] According to some embodiments of the method for assessing equivalent PM2.5 exposure concentration for urban residents in this application, obtaining the second equivalent exposure concentration value includes the following steps:

[0020] Based on the proportion of residents in city i who own and use air purification devices and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent ownership and use rate of air purification devices in city i is obtained, which is expressed by the following formula:

[0021] ACR i =ACR i *ATTR i

[0022] Among them: ACR i This indicates the proportion of residents in city i who own and use air purification devices;

[0023] Based on the average PM2.5 concentration on day t of a year in city i, the equivalent outdoor time ratio for people in city i considering air pollution forecast information, the average indoor PM2.5 infiltration attenuation rate in city i, the equivalent ownership and use rate of air purifiers in city i, the average PM2.5 purification rate of air purifiers, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. HAP，i It is expressed by the following formula:

[0024]

[0025] Among them: IEPE HAP，i This represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i, expressed in PM2.5. i，t ODR represents the average PM2.5 concentration on day t of city i in a year. i DH represents the proportion of time people in city i spend outdoors when considering those who pay attention to air pollution forecasts. i ACR represents the rate of reduction of average outdoor PM2.5 infiltration into indoor spaces in city i. i The equivalent ownership and use rate of air purifiers in city i, where DAC represents the average PM2.5 purification rate of air purifiers, and AQI. i，t DM represents the average air quality index value of city i on day t of a year, DM represents the PM2.5 penetration rate of the mask, T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is light pollution.

[0026] According to some embodiments of the method for assessing the equivalent PM2.5 exposure concentration of urban residents in this application, the method for obtaining the annual comprehensive PM2.5 equivalent exposure concentration value of urban resident i is expressed by the following formula:

[0027] IEPE i =IEPE AAP，i +IEPE HAP，i

[0028] Among them: IEPEH AP，i IEPE represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i. HAP，i This represents the annual equivalent indoor PM2.5 exposure concentration for residents of city i.

[0029] According to the method for assessing the equivalent PM2.5 exposure concentration of urban residents based on some embodiments of this application, the ATTR is obtained. i MR i CODR i ACR i and ODRi The methods include:

[0030] Information was collected through a questionnaire survey, which included the level of concern about air pollution, whether people would wear masks or cancel going out in polluted weather, whether air purification equipment was available in workplaces and residences, and the average number of hours of outdoor activities per day.

[0031] ATTR was obtained based on the information collected from the questionnaire survey. i MR i CODR i ACR i and ODR i Data information;

[0032] The transfer learning method is used to align the statistical features of the data, and a supervised machine learning model is built and trained using the transformed source domain data. The trained model is then used to predict city-level data information, and the predicted ATTR of city i is obtained. i MR i CODR i ACR i and ODR i Data information;

[0033] in:

[0034] The survey data on air pollution concern levels included categories such as almost never, occasionally, moderately, frequently, and almost daily. Concerns described as frequently and almost daily were assigned a value of 1, while those described as otherwise assigned a value of 0. The aggregated results of the indicators were used to calculate the proportion of concern for group i, and the average value was taken for each group. This yielded the ATTR (Attention Rate of Residents of City i) indicating the proportion of residents in city i who paid attention to air pollution forecast information. i ;

[0035] The survey collected information including whether residents would wear masks or cancel outdoor activities during polluted weather, and whether air purification equipment was available in their workplaces and residences. The information was assigned a value of 1 for "yes" and 0 for "no". The indicators were aggregated by group to form a ratio indicator, yielding the proportion (MR) of residents in city i who wore masks during outdoor activities during polluted weather. i The percentage of residents in city i who canceled outdoor activities during periods of air pollution (CODR) i The proportion of residents in city i who own and use air purification equipment (ACR) i ;

[0036] The survey collected information including the average number of hours of outdoor activity per day for individuals. Based on this, the proportion of daily outdoor activity for individuals or groups (i) was calculated to obtain the average daily outdoor time ratio (ODR) of residents of city i.i .

[0037] In a second aspect, the urban residents' equivalent PM2.5 exposure concentration assessment device according to some embodiments of this application includes...

[0038] First equivalent exposure concentration value acquisition module: used to acquire the annual equivalent PM2.5 exposure concentration value of urban residents' outdoor environment based on the annual monitoring data of urban residents' outdoor environment, including the first equivalent exposure concentration value;

[0039] The second equivalent exposure concentration value acquisition module is used to acquire the annual equivalent PM2.5 exposure concentration value of urban residents' indoor environment based on the annual indoor environmental monitoring data of urban residents, including the second equivalent exposure concentration value.

[0040] The module for obtaining the annual comprehensive PM2.5 equivalent exposure concentration value is used to obtain the annual comprehensive PM2.5 equivalent exposure concentration value of urban residents based on the first equivalent exposure concentration value and the second equivalent exposure concentration value.

[0041] According to some embodiments of the present application, the urban residents' equivalent PM2.5 exposure concentration assessment device obtains the first equivalent exposure concentration value by the first equivalent exposure concentration value acquisition module in the following manner:

[0042] Based on the proportion of daily average outdoor time spent by residents of city i, the proportion of residents of city i who cancel their outdoor activities during periods of air pollution, and the proportion of residents of city i who pay attention to air pollution forecasts, the equivalent outdoor time ratio (ODR) for city i, considering those who pay attention to air pollution forecasts, is obtained. i It is expressed by the following formula:

[0043] ODR i =ODR i *(1-CODR i *ATTR i )

[0044] Among them: ODR i CODR represents the proportion of daily outdoor time spent by residents of city i. i ATTR indicates the percentage of residents in city i who cancel their outdoor activities during periods of air pollution. i This indicates the percentage of residents in city i who pay attention to air pollution forecast information;

[0045] Based on the proportion of residents in city i who wear masks during outdoor activities during periods of air pollution and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent outdoor mask-wearing rate MR of city i is obtained. i It is expressed by the following formula:

[0046] MR i=MR i *ATTR i

[0047] Among them: MR i This indicates the percentage of residents in city i who wear masks during outdoor activities when air pollution is present.

[0048] Based on the average PM2.5 concentration on day t of a year in city i, the proportion of people in city i who spend the equivalent time outdoors when considering air pollution forecast information, the equivalent mask-wearing rate outdoors in city i, the PM2.5 penetration rate of masks, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. AAP，i It is expressed by the following formula:

[0049]

[0050] Among them: IEPE AAP，i This represents the annual equivalent PM2.5 exposure concentration for residents of city i in the outdoor environment. i，t ODR represents the average PM2.5 concentration on day t of city i in a year. i MR represents the equivalent outdoor time proportion of people in city i when considering those who pay attention to air pollution forecast information. i DM represents the equivalent mask-wearing rate in city i, DM represents the PM2.5 penetration rate of the mask, and AQI represents the PM2.5 penetration rate. i，t Let represent the average air quality index (AQI) value of city i on day t of a year, where T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is lightly polluted.

[0051] According to some embodiments of the urban residents' equivalent PM2.5 exposure concentration assessment device of this application, the second equivalent exposure concentration value acquisition module acquires the second equivalent exposure concentration value based on the following method:

[0052] Based on the proportion of residents in city i who own and use air purification devices and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent ownership and use rate of air purification devices in city i is obtained, which is expressed by the following formula:

[0053] ACR i =ACR i *ATTR i

[0054] Among them: ACR i This indicates the proportion of residents in city i who own and use air purification devices;

[0055] Based on the average PM2.5 concentration on day t of a year in city i, the equivalent outdoor time ratio for people in city i considering air pollution forecast information, the average indoor PM2.5 infiltration attenuation rate in city i, the equivalent ownership and use rate of air purifiers in city i, the average PM2.5 purification rate of air purifiers, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. HAP，i， It can be expressed by the following formula:

[0056]

[0057] Among them: IEPE HAP，i This represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i. i，t ODR represents the average PM2.5 concentration on day t of city i in a year. i DH represents the proportion of time people in city i spend outdoors when considering those who pay attention to air pollution forecasts. i ACR represents the rate of reduction of average outdoor PM2.5 infiltration into indoor spaces in city i. i The equivalent ownership and use rate of air purifiers in city i, where DAC represents the average PM2.5 purification rate of air purifiers, and AQI. i，t DM represents the average air quality index value of city i on day t of a year, DM represents the PM2.5 penetration rate of the mask, T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is light pollution.

[0058] According to some embodiments of the present application, the urban residents' equivalent PM2.5 exposure concentration assessment device, the annual comprehensive PM2.5 equivalent exposure concentration value acquisition module acquires the annual comprehensive PM2.5 equivalent exposure concentration value of urban residents i based on the following method:

[0059] IEPEi = IEPE AAP，i +IEPE HAP，i

[0060] Among them: IEPE HAP，i IEPE represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i. HAP，i This represents the annual equivalent indoor PM2.5 exposure concentration for residents of city i.

[0061] According to some embodiments of this application, the urban residents' equivalent PM2.5 exposure concentration assessment device obtains the ATTR based on the following method. i MR i CODR iACR i and ODR i :

[0062] Information was collected through a questionnaire survey, which included the level of concern about air pollution, whether people would wear masks or cancel going out in polluted weather, whether air purification equipment was available in workplaces and residences, and the average number of hours of outdoor activities per day.

[0063] ATTR was obtained based on the information collected from the questionnaire survey. i MR i CODR i ACR i and ODR i Data information;

[0064] The transfer learning method is used to align the statistical features of the data, and a supervised machine learning model is built and trained using the transformed source domain data. The trained model is then used to predict city-level data information, and the predicted ATTR of city i is obtained. i MR i CODR i ACR i and ODR i Data information;

[0065] in:

[0066] The survey data on air pollution concern levels included categories such as almost never, occasionally, moderately, frequently, and almost daily. Concerns described as frequently and almost daily were assigned a value of 1, while those described as otherwise assigned a value of 0. The aggregated results of the indicators were used to calculate the proportion of concern for group i, and the average value was taken for each group. This yielded the ATTR (Attention Rate of Residents of City i) indicating the proportion of residents in city i who paid attention to air pollution forecast information. i ;

[0067] The survey collected information including whether residents would wear masks or cancel outdoor activities during polluted weather, and whether air purification equipment was available in their workplaces and residences. The information was assigned a value of 1 for "yes" and 0 for "no". The indicators were aggregated by group to form a ratio indicator, yielding the proportion (MR) of residents in city i who wore masks during outdoor activities during polluted weather. i The percentage of residents in city i who canceled outdoor activities during periods of air pollution (CODR) i The proportion of residents in city i who own and use air purification equipment (ACR) i ;

[0068] The survey collected information including the average number of hours of outdoor activity per day for individuals. Based on this, the proportion of daily outdoor activity for individuals or groups (i) was calculated to obtain the average daily outdoor time ratio (ODR) of residents of city i. i .

[0069] This application also provides an electronic device, which includes: one or more processors, a memory, and one or more programs; wherein the one or more programs are stored in the memory, and the one or more programs include instructions that, when executed by the electronic device, cause the electronic device to execute the technical solution of this application.

[0070] This application also provides a computer-readable storage medium, which includes a computer program that, when run on an electronic device, causes the electronic device to execute the technical solution of this application.

[0071] Beneficial Effects: This invention, in determining the PM2.5 exposure concentration of urban residents, incorporates the interaction between air pollution risk information and protective behaviors, resulting in a more accurate representation of the actual situation. Furthermore, through machine learning and transfer learning methods, the data can be extrapolated to prefecture-level cities. For the technical effects achievable in the above aspects, please refer to the description of the various possible solutions for the first aspect above; they will not be repeated here. Additional aspects and advantages of this invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0072] Figure 1 This diagram illustrates the relationship between public awareness, perception, and concern about air pollution and their protective behaviors.

[0073] Figure 2 This is a diagram illustrating the impact of air pollution risk information on human PM2.5 exposure.

[0074] Figure 3 This is a map showing the distribution of PM2.5 exposure concentrations.

[0075] Figure 4 This is a statistical information of an important indicator in the questionnaire.

[0076] Figure 5 This is a diagram of the model training and prediction process.

[0077] Figure 6 Density distribution of predicted index values.

[0078] Figure 7 This is a flowchart of the method for assessing the equivalent PM2.5 exposure concentration of urban residents according to the present invention. Detailed Implementation

[0079] The embodiments of this application are described in detail below with reference to the accompanying drawings, examples of which are shown in the drawings. This application provides a method and apparatus to solve the problem of improving the accuracy of obtaining PM2.5 exposure concentrations of urban residents, so as to provide accurate data support for research on the relationship between air pollution and health levels. The method and apparatus, electronic devices, and computer storage media are based on the same technical concept. Since the principles of the method and apparatus, electronic devices, and computer storage media in solving the problem are similar, the implementation of the apparatus, electronic devices, and computer storage media can refer to each other, and repeated details will not be repeated.

[0080] Definition: Fine particulate matter, also known as fine particles, PM2.5, refers to particulate matter in ambient air with an aerodynamic equivalent diameter of 2.5 micrometers or less. It can remain suspended in the air for a relatively long time, and a higher concentration indicates more severe air pollution. Although PM2.5 is only a small component of the Earth's atmosphere, it has a significant impact on air quality and visibility. Compared to coarser atmospheric particles, PM2.5 particles are smaller, have a larger surface area, are more reactive, and easily carry toxic and harmful substances (e.g., heavy metals, microorganisms). Furthermore, they remain in the atmosphere for a longer time and travel greater distances, thus having a greater impact on human health and atmospheric environmental quality. In February 2013, the National Committee for Terminology in Science and Technology officially named PM2.5 "fine particulate matter" in Chinese. The chemical composition of fine particulate matter mainly includes organic carbon (OC), elemental carbon (EC), nitrates, sulfates, ammonium salts, and sodium salts (Na). + )wait.

[0081] According to risk perception theory, people's risk protection behavior is a dynamic process. An individual's level of awareness of air pollution risks is mainly influenced by family, socio-cultural factors, and prior experience. When a relevant risk event is experienced again, hypotheses about that event are activated, further generating the intention and behavior to protect oneself. When individuals or groups take protective measures against air pollution, it is equivalent to reducing their exposure to PM2.5 and other particulate matter in polluted environments. Factors affecting PM2.5 exposure mainly include the temporal / spatial (indoor / outdoor) distribution and concentration of ambient PM2.5, an individual's awareness of air pollution risk information, and their willingness to take protective measures. This invention quantifies these factors using air quality data and over 1000 questionnaires, indicating that the public's awareness of air pollution is relatively comprehensive. Figure 1 The study found a significant positive correlation between people's awareness, perception, and concern about air pollution and their protective behaviors.

[0082] according to Figure 1This chart shows the relationship between public awareness, perception, and concern about air pollution and their protective behaviors. Data points can be aggregated by region (e.g., province). Awareness and perception are represented by values ​​from 0 to 10, while concern represents the proportion of people in the survey who frequently or almost daily pay attention to air pollution. Among protective behaviors, canceling outdoor activities, wearing masks, and using air purifiers (at home / workplace) are the percentages of people who would take these actions during polluted weather.

[0083] Based on the above results and analysis, this invention establishes a comprehensive equivalent PM2.5 exposure model (IEPEM) to demonstrate the differences in PM2.5 exposure concentrations among different population groups during polluted weather. The survey questionnaire of this invention shows that approximately 82.6% of respondents obtain current air pollution levels through air quality forecasts from media such as the internet and television. Based on their sensitivity to air pollution issues such as smog, they can be divided into highly sensitive and low-sensitive groups. The highly sensitive group pays close attention to air pollution weather information and takes corresponding protective measures, such as canceling or reducing outdoor time, wearing masks, and purchasing and using air purifiers. The low-sensitive group is almost unaffected by air pollution information. For the highly sensitive group, the exposure concentration is equivalent to the ambient PM2.5 concentration attenuated by protective factors (POH, POM, POAC), while the low-sensitive group is directly exposed to ambient PM2.5 when outdoors, and is only protected by buildings indoors.

[0084] To systematically study the changes in PM2.5 exposure concentrations and their gains and losses caused by differences in the behaviors of different groups under the influence of daily air pollution risk information, this invention proposes the following five IEPEM scenarios:

[0085] Table 1 IEPEM Scenarios and Settings

[0086]

[0087] Comprehensive Equivalent PM2.5 Exposure Concentration: This invention extrapolates indicators such as residents' concern about air pollution and protective behaviors in 294 Chinese cities using questionnaire data and city / provincial statistical data, employing machine learning and other technologies. Using these indicators, the equivalent PM2.5 concentration under scenarios S0-S4 is simulated. The spatial distribution of the baseline scenario S0 indicates that air pollution mainly occurs in concentrated areas. Under scenarios S1-S4, the equivalent PM2.5 concentration decreases to varying degrees compared to scenario S0, and the degree of change in PM2.5 concentration compared to S0 varies across different regions under different scenarios. This indicates that differences in population activity patterns and protective behaviors in different regions directly impact PM2.5 exposure concentrations.

[0088] from Figure 3It can be seen that the average PM2.5 concentration in the environment under scenario S0 is 34.1 μg / m3, and the comprehensive equivalent PM2.5 exposure concentrations under scenarios S1-S4 are 23.3 μg / m3, respectively. 3 20.3 μg / m 3 21.8 μg / m 3 19.8 μg / m 3 Compared to scenario S0, the levels decreased by 10.8 μg / m³. 3 13.8 μg / m 3 12.3 μg / m 3 14.3 μg / m 3 Among them, S4 showed the largest decrease, indicating that when the public pays attention to air pollution information and takes protective measures, changing air pollution risk information can significantly affect their overall PM2.5 exposure concentration. At the same time, it can be seen that among all protective factors, building protection plays the most important role in the reduction of environmental PM2.5 exposure concentration. This is because people spend about 80% of their time indoors, and buildings have a more significant effect on blocking external PM2.5. Compared with the settings of scenarios S0-S2, S3 is closer to the real-world scenario. Under the assumption that protective measures are only taken when people pay attention to air pollution, the overall PM2.5 exposure concentration increased by 1.5 μg / m3 compared to the ideal scenario (S2).

[0089] Differences in people's activity patterns across different regions significantly affect the per capita PM2.5 exposure concentration in those regions. People's awareness and behavior regarding protection are higher in areas with more severe air pollution than in areas with less pollution. Public awareness and behavior are not only influenced by the actual level of risk, but also highly correlated with differences in risk information perception and understanding caused by uneven economic and cultural development across different regions.

[0090] This invention conducted a detailed survey of the Chinese public's perception, awareness, and concern about air pollution, as well as their protective behaviors during polluted weather, using questionnaires. By employing machine learning and transfer learning methods, this "small data" survey was extrapolated to prefecture-level cities across the country, revealing a comprehensive picture of how public behavior patterns change under air pollution and uncovering the benefits of protective behaviors influenced by environmental pollution information. The study shows that ignoring people's activity patterns significantly overestimates the harmfulness of PM2.5, and that proactive protective behaviors bring significant health benefits. Incorporating indoor and outdoor activity patterns into the estimation model makes it more realistic when considering protective behaviors. Furthermore, differences in protective awareness and behaviors among different genders, ages, and urban / rural areas create gaps in benefits and losses. Further research indicates that information inequality is a significant reason for regional differences in PM2.5-related benefits and losses.

[0091] In one embodiment, the present invention designed a questionnaire to obtain data on awareness and protection against air pollution from different regions and groups. The questionnaire underwent rigorous quality control, eliminating questionnaires with obvious logical errors, missing data, and inconsistent addresses. Ultimately, 1072 valid questionnaires were obtained from 1209 responses. Initial statistical information for some important indicators in the questionnaire is as follows: Figure 4 As shown. Perceptions and understanding of air pollution were primarily assessed by asking respondents about their perceived severity of air pollution in their area and its impact on health, expressed as a numerical value from 1 to 10, with higher numbers indicating greater severity. Values ​​greater than 5 were considered relatively high. Attention to air pollution was categorized as: almost never - occasionally - moderately - frequently - almost daily. This invention uses the aggregated results of this index, assigning a value of 1 to "frequent" or higher attention levels, and 0 to others. The average of these values ​​was used to obtain the attention ratio (ATTRi) for group i (e.g., region, gender, age, etc.). Air pollution weather conditions are... Indicators such as whether or not to wear masks or cancel going out, and whether or not air purification equipment is available in workplaces and residences are used. A value of 1 is assigned to the answer "yes", and 0 is assigned to the answer "no". In this invention, this indicator is also used as a ratio indicator after the aggregation of groups, namely, the mask wearing rate (MRi), the cancellation rate of going out (CODRi), and the indoor air purification equipment ownership rate (ACRi, the average of the air purification equipment ownership rates in workplaces and residences) of group i. The outdoor activity time ratio mainly investigates the average number of hours of outdoor activity per day for individuals, and then calculates the daily outdoor activity ratio (ODRi) of individuals or group i.

[0092] Figure 4 Statistical information on key indicators in the survey questionnaire. Figure 4 A represents whether or not one will wear a mask when going out after receiving an air pollution weather warning or forecast; Figure 4 B refers to whether one will cancel going out when receiving an air pollution weather warning or forecast. Figure 4 C refers to whether air purification equipment is installed in the workplace; Figure 4 D indicates whether air purification equipment is installed in the residential area; Figure 4 The value of E is whether or not one pays attention to air pollution early warning and forecast information in daily life; Figure 4 The letter F indicates the air quality level at which protective measures and actions will be taken. Figure 4 G represents the distribution of outdoor activity hours across different intervals, with the vertical axis representing the proportion of each interval and the horizontal axis representing the time interval.

[0093] As is well known, data obtained from questionnaires is generally limited. Previous studies on air pollution-related protective behaviors have typically focused on a single city or used a small, statistically significant sample to represent the entire population of each province. While this approach has some merit, it is clearly crude and cannot accurately reflect the high heterogeneity among different cities and regions. Since weather and air quality forecasts are generally aggregated at the city level, and residents' perceptions and awareness of air pollution vary across cities, protective behaviors in this invention should also be city-level. However, questionnaire data clearly cannot cover all cities, or its sample size is not statistically significant. When more than 20 samples are considered statistically significant, over 20 provinces and regions in the questionnaires used in this invention meet this standard, but fewer than 10 cities meet or exceed it. A feasible solution is to use provincial-level questionnaire results as the predictable items, use other provincial statistical data as auxiliary features to build a machine learning prediction model, and then use city-level statistical data of similar significance as model input to predict city-level attention and protective behaviors. However, this method still has potential drawbacks: the joint probability distributions of city-level data and provincial-level data may differ, which does not conform to the principle of identically and independently distributed (iiD) training and prediction data in machine learning models.

[0094] To address the above issues, this invention introduces transfer learning into the inference of city-level protective behaviors and attention levels. The idea behind transfer learning is to leverage the similarity of data, tasks, or models to apply models and knowledge learned in an old domain to a new domain. This can be uniformly represented in the following form:

[0095]

[0096] N in formula 1 s v is the number of samples in the source domain. i Let D be the source domain sample weights, l(·,·) be the empirical loss function, R(·,·) be the transfer regularization term, and T(·,·) be the expression applied to the source domain D. s and target domain D tThe feature transformation function of the sample. Transfer learning can be divided into sample weight transfer method, feature transformation transfer method, and pre-trained model transfer method according to different parts of the optimization formula (1). The transfer learning method used in this invention is the subspace feature transformation method. This method usually assumes that the source domain and target domain data will have similar distributions in the transformed subspace. By aligning the statistical features of the data, the data distribution is aligned. It is suitable for unsupervised transfer without labeling in the target domain and is more in line with the extrapolation task of this invention. The CORAL (Correlation Alignment) algorithm is a relatively mature method in the subspace transformation transfer method and is also the specific method applied in this invention. The CORAL algorithm mainly aligns the second-order features of the two domains to achieve the alignment of feature data distribution. It assumes that C s C t Let A be the covariance matrices of the source and target domains, respectively. Transforming A into the objective function, the optimization objective is:

[0097]

[0098] Based on the problem and data of this invention, the calculation process for the final required prediction result is as follows:

[0099] Step 1: Align the provincial statistical data (Source domain) with the city statistical data (Target domain).

[0100]

[0101]

[0102] C s =∑ s +eye(m) (5)

[0103] C t =∑ t +eye(m) (6)

[0104]

[0105] Equations (3)-(4) represent the feature datasets of the source and target domains, respectively. Let m be the m-th feature of a single sample in the dataset. The source domain feature data consists of provincial-level statistical data from the China Statistical Yearbook 202048, and the target domain feature data consists of urban statistical data from the China Urban Statistical Yearbook 202049. Both the source and target domain features are statistical indicators of the same type, encompassing 18 indicators across areas such as economy, environment, education, and population structure. It is noted that these indicators have significantly different data scales at the city and provincial levels. Therefore, this invention divides all indicators by the current total population of the region to obtain the per capita value of each indicator, thus mitigating prediction errors caused by this scale difference.

[0106] Step 2: Build and train a supervised machine learning model using the transformed source domain data, and finally use the trained model to predict city-level data.

[0107] Model architecture such as Figure 5 As shown in the figure. The dataset, consisting of five indicators, forms the training datasets for the five tasks. Due to the relatively small amount of data in this invention, overfitting is prone to occur, making it unsuitable to use highly complex models. Therefore, this invention selects four models as candidate models: Random Forest (RF), Lasso Regression, Ridge Regression, and Support Vector Machine (SVM / R). These models are simple in structure and efficient, and overfitting can be easily limited using regularization techniques. During training, a hyperparameter grid search method is used to automatically optimize the training for each task. Five-fold cross-validation is used to verify the accuracy of each training result, selecting the best-performing model for each task. Finally, city-level sample data is used to predict the corresponding indicators for each city.

[0108] Figure 5 The model training and prediction process is shown in Table 2. The training and testing results are also shown in Table 2. Column 1 lists the various metrics, column 2 lists the corresponding best model, and columns 3 and 4 show the mean squared error (MSE) and goodness of fit (R²) of the 5-fold cross-validation under the optimal parameters of the corresponding model, respectively. 2 The values, as can be seen from the corresponding measurement data, show that the model maintained a good fit. Column 5 shows the test results in cities with a sample size greater than 20, with the mean absolute error (MAE) as the metric. It can be seen that the predicted results for most indicators deviate from the actual data by about 0.1, which is within an acceptable range.

[0109] Figure 6 The distribution chart shows the prediction results for the five indicators. Some indicators have very few outliers. To make the final results more stable and reliable, this invention trims all values ​​that exceed the mean μ of each indicator by ±2 standard deviations δ to μ±2δ.

[0110] Table 2. Model Training and Testing Results

[0111] Best Model MSE(CV) R-Square (CV) MAE(TEST) <![CDATA[ATTR s,i ]]> Lasso 0.0063 0.7136 0.1009 <![CDATA[ODR s,i ]]> Ridge 0.0046 0.7365 0.0408 <![CDATA[CODR s,i ]]> RF 0.0053 0.7900 0.0722 <![CDATA[MR s,i ]]> Lasso 0.0049 0.7184 0.1394 <![CDATA[ACR s,i ]]> Lasso 0.0129 0.6516 0.1082

[0112] Figure 6 The table shows the density distribution of predicted index values. When categorizing by age, gender, and urban / rural groups, the sample size for each group decreases further within each city group due to indicators such as awareness and protective behavior, making effective inferences impossible. Therefore, the index values ​​for each group are calculated using the total original questionnaire data, as shown in Table 3. Since air purification equipment can be shared by all groups in the same space, the ACR indexes for gender, age, and other groups are still calculated based on the overall predicted values ​​for each city, while the urban / rural group data is based on the original questionnaire statistics.

[0113] Table 3. Values ​​of each indicator for different groups

[0114] Group ODR MR CODR ATTR ACR female 0.210 0.788 0.713 0.428 No male 0.245 0.756 0.611 0.413 No rural 0.252 0.753 0.634 0.354 0.473 city 0.218 0.779 0.676 0.441 0.620 age_0_14 0.227 0.736 0.566 0.208 No age_15_64 0.226 0.773 0.670 0.435 No age_65 0.223 0.764 0.706 0.525 No

[0115] Equivalent PM2.5 Calculation: The equivalent PM2.5 exposure concentration calculation model of this invention references and improves upon the Integrated Population Weighted Exposure (IPWE) model developed by Shen et al. The IPWE model divides air pollution into indoor air pollution (HAP) and outdoor ambient air pollution (AAP), with HAP mainly originating from household solid fuel use. The model also uses the proportion of time people spend in different spaces as the weight for summing the PM2.5 concentration in each space, as shown in Formula 8.

[0116]

[0117] Where C0 represents the outdoor PM2.5 concentration, t m Let C be the time C spends in space m. m Let m be the PM2.5 concentration in space. Compared to many earlier studies that directly used the ambient PM2.5 concentration as the PM2.5 exposure concentration for residents, the IPWE model is more reasonable and accurate, and therefore widely used. However, this model still has the following shortcomings: the model equates indoor PM2.5 concentration to the sum of outdoor ambient PM2.5 concentration and PM2.5 concentration generated by indoor pollution sources, but many studies have shown that outdoor PM2.5 cannot enter indoors without obstruction and will have a significant attenuation effect. Therefore, this invention incorporates the PM2.5 penetration coefficient of various regions into the improved model; more generally, no studies have been found to include human protective behaviors, such as wearing masks in smoggy weather, canceling outdoor activities, purchasing and using air purification equipment, etc., which all have a direct impact on human PM2.5 exposure concentration. Based on the above, this invention establishes the Integrated Equivalent PM2.5 Exposure Model (IEPEM) to more accurately measure people's actual PM2.5 exposure concentration. To construct this model, this invention first makes the following assumptions:

[0118] Hypothesis 1: The activity patterns, protective behaviors, and attention behaviors of each city and group all conform to a normal distribution and are independent of each other.

[0119] Hypothesis 2: People will only decide whether to take protective measures after receiving information about air pollution.

[0120] Assumption 3: People will only take protective measures when air pollution reaches a certain level. This invention selects "light pollution" as the air pollution level, meaning the Air Quality Index (AQI) is greater than the minimum AQI value corresponding to this level.

[0121] Based on the above assumptions, the IEPEM model can be expressed by Equation 9. The proportion of urban attention and the ratio of protective behaviors in the equation are derived from urban protection data and the prediction results of the inferred portion, and both follow N(μ) i θ 2 ) distribution, μ i Here is the predicted data for this indicator in city i, and θ is the standard deviation of this indicator. i，t The average PM2.5 concentration in city i on day t is represented by the Air Quality Index (AQI), which is derived from data from over 2000 ground-based air quality monitoring stations under the Ministry of Ecology and Environment of China. i，t IEPE represents the air quality index of city i on day t. i DH represents the annual comprehensive equivalent PM2.5 exposure value for city i. Threshold is the AQI value at which the air quality level of "light pollution" is reached. DM represents the protective effect of the mask, i.e., the attenuation rate of PM2.5 after filtration by the mask; the protective effect follows the Chinese government's group standard F9054 for "PM2.5 Protective Masks". i The attenuation rate represents the protective effect of buildings in different regions, i.e., the rate attenuation of outdoor PM2.5 when it penetrates into the indoor environment. DAC represents the purification efficiency of air purifiers, i.e., the rate attenuation of PM2.5 concentration in indoor air after purification by air purifiers, using the mean of these studies as the attenuation rate value. To account for uncertainty, it is assumed that all types of attenuation rate data follow a normal distribution with a standard deviation of 0.1.

[0122]

[0123] in:

[0124] Based on the proportion of daily average outdoor time spent by residents of city i, the proportion of residents of city i who cancel their outdoor activities during periods of air pollution, and the proportion of residents of city i who pay attention to air pollution forecasts, the equivalent outdoor time ratio (ODR) for city i, considering those who pay attention to air pollution forecasts, is obtained. i It is expressed by the following formula:

[0125] ODR i =ODR i*(1-CODR i *ATTR i )

[0126] Among them: ODR i CODR represents the proportion of daily outdoor time spent by residents of city i. i ATTR indicates the percentage of residents in city i who cancel their outdoor activities during periods of air pollution. i This indicates the percentage of residents in city i who pay attention to air pollution forecast information;

[0127] Based on the proportion of residents in city i who wear masks during outdoor activities during periods of air pollution and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent outdoor mask-wearing rate MR of city i is obtained. i It is expressed by the following formula:

[0128] MR i =MR i *ATTR i

[0129] Among them: MR i This indicates the percentage of residents in city i who wear masks during outdoor activities when air pollution is present.

[0130] Based on the proportion of residents in city i who own and use air purification devices and the proportion of residents in city i who pay attention to air pollution forecast information, the equivalent ownership and use rate of air purification devices in city i is obtained, which is expressed by the following formula:

[0131] ACR i =ACR i *ATTR i

[0132] Among them: ACR i This indicates the proportion of residents in city i who own and use air purification devices;

[0133] Based on the average PM2.5 concentration on day t of a year in city i, the proportion of people in city i who spend the equivalent time outdoors when considering air pollution forecast information, the equivalent mask-wearing rate outdoors in city i, the PM2.5 penetration rate of masks, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. AAP，i， It can be expressed by the following formula:

[0134]

[0135] Among them: IEPE AAP，i This represents the annual equivalent PM2.5 exposure concentration for residents of city i in the outdoor environment. i，tODR represents the average PM2.5 concentration on day t of city i in a year. i MR represents the equivalent outdoor time proportion of people in city i when considering those who pay attention to air pollution forecast information. i DM represents the equivalent mask-wearing rate in city i, DM represents the PM2.5 penetration rate of the mask, and AQI represents the PM2.5 penetration rate. i，t Let represent the average air quality index (AQI) value of city i on day t of a year, where T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is lightly polluted.

[0136] Based on the average PM2.5 concentration on day t of a year in city i, the equivalent outdoor time ratio for people in city i considering air pollution forecast information, the average indoor PM2.5 infiltration attenuation rate in city i, the equivalent ownership and use rate of air purifiers in city i, the average PM2.5 purification rate of air purifiers, and the average air quality index on day t of a year in city i, the annual equivalent PM2.5 exposure concentration (IEPE) for residents of city i is obtained. HAP，i It is expressed by the following formula:

[0137]

[0138] Among them: IEPE HAP，i This represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i. i，t ODR represents the average PM2.5 concentration on day t of city i in a year. i DH represents the proportion of time people in city i spend outdoors when considering those who pay attention to air pollution forecasts. i ACR represents the rate of reduction of average outdoor PM2.5 infiltration into indoor spaces in city i. i The equivalent ownership and use rate of air purifiers in city i, where DAC represents the average PM2.5 purification rate of air purifiers, and AQI. i，t DM represents the average air quality index value of city i on day t of a year, DM represents the PM2.5 penetration rate of the mask, T represents the number of days in the year, t represents the day t of the year, and Threshold represents the AQI value of city i when the air quality level is light pollution.

[0139] The annual equivalent exposure concentration of PM2.5 for residents of city i is obtained by the following formula:

[0140] IEPE i =IEPE AAP，i +IEPE HAP，i

[0141] Among them: IEPE HAP，iIEPE represents the annual equivalent PM2.5 exposure concentration in the indoor environment for residents of city i. HAP，i This represents the annual equivalent indoor PM2.5 exposure concentration for residents of city i.

[0142] Table 4 shows the different indicator settings under scenarios S0-S4. "Yes" indicates that the actual value of the indicator is retained, while "No" indicates that the indicator is not considered. An AQI of 100 or 35 represents the threshold for "light pollution" according to ambient air quality standards and national air quality standards, respectively.

[0143] Table 4. Indicator settings under different scenarios

[0144] Index S0 S1 S2 S3 S4 <![CDATA[ATTR s，i ]]> 0 0 1 Yes Yes <![CDATA[ODR s，i ]]> 1 Yes Yes Yes Yes <![CDATA[CODR s，i ]]> 0 0 Yes Yes Yes <![CDATA[MR s，i ]]> 0 0 Yes Yes Yes <![CDATA[ACR s，i ]]> 0 No Yes Yes Yes Threshold No No 100 100 35

[0145] The electronic device provided in this application embodiment can be implemented on an electronic device with the following hardware architecture, as all the technical solutions involved in the above embodiments can be implemented on such a device. Specifically, the electronic device includes a processor, an external memory interface, an internal memory, a universal serial bus interface, a charging management module, a power management module, a battery, an antenna, a mobile communication module, a wireless communication module, a sensor module, an indicator, a camera, a display screen, and a user identification module card interface, etc.

[0146] A processor may include one or more processing units. For example, a processor may include an application processor, a modem processor, a graphics processor, an image signal processor, a controller, a memory, a video codec, a digital signal processor, a baseband processor, and a neural network processor. Different processing units may be independent devices or integrated into one or more processors. The controller may serve as the nerve center and command center of the electronic device. The controller can generate operation control signals based on instruction opcodes and timing signals, and perform instruction fetching and execution control.

[0147] A processor may include one or more interfaces.

[0148] Wireless communication functionality in electronic devices can be achieved through antennas, mobile communication modules, wireless communication modules, modem processors, and baseband processors. Antennas are used to transmit and receive electromagnetic wave signals. Each antenna in an electronic device can cover one or more communication frequency bands.

[0149] A mobile communication module can provide wireless communication solutions for electronic devices. A mobile communication module may include at least one filter, switch, power amplifier, low-noise amplifier, etc. The mobile communication module can receive electromagnetic waves via an antenna, filter and amplify the received electromagnetic waves, and then transmit them to a modem processor for demodulation. The mobile communication module can also amplify the signal modulated by the modem processor and radiate it as electromagnetic waves via the antenna. In some embodiments, at least some functional modules of the mobile communication module can be housed within a processor.

[0150] Wireless communication modules can provide solutions for wireless communication applications in electronic devices, including wireless local area networks, Bluetooth, global navigation satellite systems, frequency modulation, short-range wireless communication technology, infrared technology, and more.

[0151] Electronic devices achieve their display functions through GPUs, displays, and application processors. A GPU is a microprocessor for image processing that connects the display and the application processor.

[0152] A display screen is used to display images, videos, etc., and includes a display panel.

[0153] Electronic devices can perform shooting functions, or image acquisition functions, through ISPs, cameras, video codecs, GPUs, displays, and application processors. Cameras are used to capture still images or videos.

[0154] Internal memory can be used to store executable program code, including instructions. Internal memory may include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc. The data storage area may store data created during the use of the electronic device, as well as shared data sent by other electronic devices. Furthermore, internal memory may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, general-purpose flash memory, etc. The processor executes various functional applications and data processing of the electronic device by running instructions stored in the internal memory and / or instructions stored in memory disposed within the processor. In some embodiments, the processor distributes shared data stored in the electronic device to other devices by running instructions stored in the internal memory.

[0155] An external storage interface (ESI) can be used to connect external memory cards, such as microSD cards, to expand the storage capacity of electronic devices. The external memory card communicates with the processor through the ESI to perform data storage functions, such as saving images and videos to the external memory card.

[0156] The memory is used to store program instructions and data. Specifically, program instructions may include program code, which includes computer operation instructions. The memory may include random access memory and may also include non-volatile memory, such as at least one disk storage device. The processor executes the program instructions stored in the memory and uses the data stored in the memory to implement the above functions, thereby implementing the method provided in the above embodiments.

[0157] It is understood that the memory in this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, or flash memory. Volatile memory can be random access memory used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory, dynamic random access memory, synchronous dynamic random access memory, double data rate synchronous dynamic random access memory, enhanced synchronous dynamic random access memory, synchronous linked dynamic random access memory, and direct memory bus random access memory. It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory. Based on the above embodiments, this application also provides a computer program that, when run on a computer, causes the computer to perform the methods provided in the above embodiments.

[0158] Based on the above embodiments, this application also provides a computer storage medium storing a computer program, which, when executed by a computer, causes the computer to perform the methods provided in the above embodiments.

[0159] The storage medium can be any available medium that a computer can access. For example, but not limited to, a computer-readable medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0160] Based on the above embodiments, this application also provides a chip for reading a computer program stored in a memory to implement the method provided in the above embodiments.

[0161] Based on the above embodiments, this application provides a computer program product that implements the methods provided in the above embodiments when the computer program product is run on an electronic device.

[0162] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0164] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0166] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for assessing the equivalent PM2.5 exposure concentration of urban residents, characterized in that, include Based on annual outdoor environmental data for urban residents, the annual equivalent PM2.5 exposure concentration for residents of city i is obtained. ; Based on annual data related to the indoor environment of urban residents, obtain urban... The annual equivalent PM2.5 exposure concentration in the indoor environment of residents ; according to and Get city The annual equivalent PM2.5 exposure concentration of residents ; in: in: In the formula: Represents city The first of the year Daily average PM2.5 concentration value, Represents city The equivalent proportion of outdoor time when considering people who pay attention to air pollution forecast information. Represents city The equivalent mask wearing rate when going out, This indicates the PM2.5 penetration rate of the mask. Represents city The first of the year Daily average air quality index value, Indicates the number of days in that year. Indicates the first year sky, Represents city The AQI value when the air quality level is lightly polluted; in: In the formula: Represents city The proportion of residents' average daily time spent outdoors Represents city The percentage of residents who cancel their outdoor activities during periods of air pollution. Represents city The percentage of residents who pay attention to air pollution forecast information; in: In the formula: Represents city The percentage of residents who wear masks when engaging in outdoor activities during periods of air pollution; in: In the formula: Represents city The average rate of reduction of outdoor PM2.5 infiltration into indoor spaces. Represents city Equivalent ownership and use rate of air purification equipment This indicates the average PM2.5 purification rate of the air purification equipment. in: In the formula: Represents city The percentage of residents who own and use air purification equipment.

2. The method for assessing equivalent PM2.5 exposure concentration for urban residents according to claim 1, characterized in that, get , , , and The methods include: Information was collected through a questionnaire survey, which included the level of concern about air pollution, whether people would wear masks or cancel going out in polluted weather, whether air purification equipment was available in workplaces and residences, and the average number of hours of outdoor activities per day. Based on the information collected from the questionnaire survey , , , and Data information; The transfer learning method is used to align the statistical features of the data, and a supervised machine learning model is built and trained using the transformed source domain data. The trained model is then used to predict city-level data information, and the predicted information for city i is obtained. , , , and Data information; in: The survey data on air pollution concern levels included categories such as almost never, occasionally, moderately, frequently, and almost daily. Concerns described as frequently and almost daily were assigned a value of 1, while those described as otherwise were assigned a value of 0. The aggregated results of the air pollution concern index were used to calculate the concern ratio for each group by taking the mean. ; The survey collected information including whether individuals would wear masks or cancel outdoor activities during periods of air pollution, and whether they had air purification equipment at their workplace and residence. The information was categorized as follows: "yes" was assigned 1, "no" was assigned 0. The indicators for whether individuals would wear masks or cancel outdoor activities, and whether they had air purification equipment at their workplace and residence, were aggregated by group to form a percentage index. , , ; The survey collected information including the average number of hours of outdoor activity per day for individuals. Based on this, the proportion of daily outdoor activity for individuals or groups was calculated. .

3. A device for assessing the equivalent PM2.5 exposure concentration of urban residents, characterized in that, include The first equivalent exposure concentration value acquisition module is used to obtain urban equivalent exposure concentration values ​​based on annual outdoor environmental monitoring data of urban residents. The annual equivalent PM2.5 exposure concentration of residents in the outdoor environment ; The second equivalent exposure concentration value acquisition module is used to obtain urban equivalent exposure concentration values ​​based on annual indoor environmental monitoring data of urban residents. The annual equivalent PM2.5 exposure concentration in the indoor environment of residents ; The module for obtaining the annual comprehensive PM2.5 equivalent exposure concentration value is used to calculate the equivalent exposure concentration value based on the environmental conditions. and Get city The annual equivalent PM2.5 exposure concentration of residents ; in: in: In the formula: represents the city The average PM2.5 concentration on day t of the year. Represents city The equivalent proportion of outdoor time when considering people who pay attention to air pollution forecast information. Represents city The equivalent mask wearing rate when going out, This indicates the PM2.5 penetration rate of the mask. Represents city The first of the year Daily average air quality index value, Indicates the number of days in that year. Indicates the first year sky, Represents city The AQI value when the air quality level is lightly polluted; in: In the formula: Represents city The proportion of residents' average daily time spent outdoors Represents city The percentage of residents who cancel their outdoor activities during periods of air pollution. Represents city The percentage of residents who pay attention to air pollution forecast information; in: In the formula: Represents city The percentage of residents who wear masks when engaging in outdoor activities during periods of air pollution; in: In the formula: represents the city The average rate of reduction of outdoor PM2.5 infiltration into indoor spaces. Represents city Equivalent ownership and use rate of air purification equipment This indicates the average PM2.5 purification rate of the air purification equipment. in: In the formula: Represents city The percentage of residents who own and use air purification equipment.

4. The device for assessing equivalent PM2.5 exposure concentration for urban residents according to claim 3, characterized in that, Obtained in the following manner , , , and : Information was collected through a questionnaire survey, which included the level of concern about air pollution, whether people would wear masks or cancel going out in polluted weather, whether air purification equipment was available in workplaces and residences, and the average number of hours of outdoor activities per day. Based on the information collected from the questionnaire survey , , , and Data information; The transfer learning method is used to align the statistical features of the data, and a supervised machine learning model is built and trained using the transformed source domain data. The trained model is then used to predict city-level data information, and the predicted information for city i is obtained. , , , and Data information; in: The survey data on air pollution concern levels included categories such as almost never, occasionally, moderately, frequently, and almost daily. Concerns described as frequently and almost daily were assigned a value of 1, while those described as otherwise were assigned a value of 0. The aggregated results of the air pollution concern index were used to calculate the concern ratio for each group by taking the mean. ; The survey collected information including whether individuals would wear masks or cancel outdoor activities during periods of air pollution, and whether they had air purification equipment at their workplace and residence. The information was categorized as follows: "yes" was assigned 1, "no" was assigned 0. The indicators for whether individuals would wear masks or cancel outdoor activities, and whether they had air purification equipment at their workplace and residence, were aggregated by group to form a percentage index. , , ; The survey collected information including the average number of hours of outdoor activity per day for individuals. Based on this, the proportion of daily outdoor activity for individuals or groups was calculated. .

5. An electronic device, the electronic device comprising: One or more processors, a memory, and one or more programs; wherein the one or more programs are stored in the memory, and the one or more programs include instructions that, when executed by the electronic device, cause the electronic device to perform the method of any one of claims 1-2.

6. A computer-readable storage medium comprising a computer program that, when executed on an electronic device, causes the electronic device to perform the method of any one of claims 1-2.

Images